Want to know if this job is worth applying to?
India
Remote only
About the Company
Why We Built Lyric: Supply chains are more critical and complex than ever. Every day, large enterprises navigate trillions of possible decisions that could impact the bottom line. Powerful algorithms and AI can address these problems, yet most organizations struggle to leverage supply chain AI at scale. The current SCM technologies are either rigid, limited-scope point solutions or custom solutions built in-house, which demand immense expertise and investment.
That is…until now.
Enter Lyric: Lyric is an enterprise AI platform built specifically for supply chains, offering the best of both worlds:
Out-of-the-box AI solutions for optimizing networks, allocating inventory, scheduling routes, planning fulfillment capacity, promising orders, propagating demand, building predictions, analyzing scenarios, and more, plus
A platform-first approach that empowers both business and technical users with end-to-end product composability, leveraging no-code tools, their own code, or even forking our code to build and refine supply chain decision intelligence
With Lyric, enterprises no longer have to choose between flexibility and speed, they get both.
The Mission: We’re building a new era in supply chain with the team best equipped to lead it. With over 20 years at the intersection of supply chain and algorithms, we developed a deep conviction that global supply chains needed something like Lyric. Since our inception in December 2021, that conviction has been validated time and time again.
Today, a growing number of Fortune 500 companies, including Smurfit WestRock, Estée Lauder, Coca-Cola, and more, are innovating on their own terms with Lyric. We can’t wait to see what our customers, both current and future, are empowered to build with us next. Come build with us!
Most AI engineering roles today are about building agents. This one is about building the framework that agents are built on.
At Lyric, we're building an agentic platform that lets both our own teams and our customers create, orchestrate and run agents in production. The work sits one abstraction layer above any individual agent: the primitives, execution loops, orchestration patterns and infrastructure that many different agentic use cases have to run on.
You'll work across both sides of that platform — the framework developers and customers build agents with, and the agents built on top of it that power experiences across Lyric, including AgentFlow and our Copilot / Ask Q experiences.
So the interesting engineering question isn't “how do we build this agent?” It's increasingly “what should the underlying system look like so that many different agents can be built reliably on top of it?” That's the problem we want you to help solve.
We're looking for Platform Engineers with real production experience building agentic systems — people who want to work close to the edge of how agentic software is actually being built and deployed today.
Build and evolve the core framework used to create and orchestrate agents across Lyric.
Design and implement production-grade agentic loops, tools, abstractions and execution patterns.
Build agents on top of that framework, not only the infrastructure underneath them.
Own meaningful engineering problems end-to-end — from figuring out the approach to shipping the implementation.
Contribute your own ideas rather than working from tightly prescribed specifications.
Evaluate emerging techniques and translate what's useful from papers, new frameworks and the broader AI ecosystem into production systems.
Work closely with a small GenAI engineering team on problems where the right implementation is often still being discovered.
Help establish engineering patterns for a category of software that is changing extremely quickly.
The most important requirement is real agentic engineering experience. You should have:
Built agentic systems or applications that run in production, rather than only personal projects or proofs of concept.
Hands-on experience with agent frameworks, orchestration, tool use, multi-step execution or similar agentic architectures.
Strong software engineering fundamentals, and a track record of taking ambiguous problems from idea to working system.
Experience writing production software in Python.
The ability to pick up unfamiliar systems independently, explore possible implementations and make sound engineering decisions.
A high degree of technical curiosity, and comfort operating in an ecosystem where best practices can change quickly.
We're broadly targeting engineers with roughly 4–6 years of software engineering experience, but depth of relevant experience matters more than the number itself.
A large part of the market currently builds with Python, LangChain, LangSmith and other contemporary agent frameworks and orchestration tooling. That's completely fine. Our own stack is Python plus significant infrastructure we've built in-house.
We don't expect you to arrive knowing our exact tools. We care much more about whether you understand the underlying concepts well enough to learn a new stack quickly.
Your background doesn't necessarily say “AI Engineer”. Some of the strongest engineers in this space started as backend, full-stack or even frontend engineers, then moved aggressively into applied AI as the ecosystem evolved.
You may fit this role if you've become the person on your team who:
keeps experimenting with new agentic techniques before they become standard;
understands what actually holds up once an agent meets production;
reads papers, technical discussions and emerging implementations because the answer isn't in the documentation yet;
builds rather than only talks about AI;
can get from an ambiguous problem to a functioning implementation without needing every step defined.
Formal ML research credentials are not required.
Strong product instincts and an ability to connect technical decisions to user outcomes.
Experience building developer platforms, frameworks or internal infrastructure.
Experience exposing agentic capabilities to other developers or external customers.
A habit of following current AI research and engineering developments, and testing promising ideas yourself.
Experience working in an environment where the technical architecture is evolving quickly.
We're looking for engineers who combine solid software fundamentals with unusually fast learning and real depth in agentic systems. If you've already been building this kind of software in production — and want to build the infrastructure that makes the next generation of it possible — we'd like to talk.
